Venous thromboembolism (VTE) accounts for 100,000-180,000 deaths, greater than 250,000 hospitalizations, and between $13,000 and $20,000 annually per event in the United States. VTE is the third leading cause of cardiovascular death after myocardial infarction and stroke. The long-term sequelae of VTE include post thrombotic syndrome and pulmonary hypertension which are each associated with low health-related quality of life (HRQOL), reduced survival, and high costs. Without prophylaxis, the prevalence of VTE following hip or knee arthroplasty is 40-60%. Given the mortality and morbidity associated with VTE, there is universal agreement that VTE prophylaxis should be given, but optimal prophylaxis is uncertain and increasingly controversial. Specialty consensus guidelines differ leading to wide variation in clinical practice. Further heightening the uncertainty, higher all-cause mortality has been reported with use of potent anticoagulants and a pooled analysis of VTE prevention trials reported a 7-fold higher risk of death among patients who sustained a major bleeding event. The Institute of Medicine has designated VTE prophylaxis following joint replacement as a priority area for comparative effectiveness research to better elucidate the benefits and harms of different antithrombotic strategies among the diverse groups of patients cared for in real-world practice. Medically complex patients, patients in the oldest age groups, and those with moderate to severe obesity are vastly under-represented in trials. In addition, few trials incorporate information about an intervention's associated long term HRQOL, survival, and cost, which are the outcomes most important to patients, providers, and policy makers. To address these concerns, we will prospectively assemble a cohort of >21,000 patients undergoing elective joint replacement from 65 community hospitals to compare the effectiveness of different regimens of VTE prophylaxis. We will use various methods (risk modeling, propensity scores, disease simulation modeling) to inform the choice of VTE prophylaxis in a population of unselected, community-based patients who are older and sicker than previously studied populations. Utilizing the unique assets of the Institute for Health Metrics, we will pursue the following aims: 1) to determine incidence of VTE, major hemorrhage, and mortality within 90 days of hip or knee arthroplasty by age and sex, 2) to develop risk-prediction models for VTE, hemorrhage, and mortality following arthroplasty, 3) to compare the effectiveness and harms associated with high intensity versus low intensity VTE prevention strategies, and 4) to determine the health quality, survival, and costs for VTE prophylaxis agents using disease simulation. Reduction of VTE is a pressing U.S. health concern. The study investigators are well-versed in the conduct of large scale studies and contribute complementary expertise. Our proposal will inform critical knowledge gaps in our understanding of the safety and efficacy of different antithrombotic strategies across the spectrum of patients in routine practice. The public health significance is enormous given the projected need for joint replacement in this patient population.